Pollute: Pollute data

In tanujitdey/modelSampler: A unique tool for variable selection and model exploration in linear regression model

Description

This data set consist of mortality rates against socio-economic, meteorological and pollution variables for 60 statistical areas in the USA. For details of the data set, see references below.

Format

The data set have 60 observations with 15 variables. “MortRate” is the outcome variable. The data frame containing:

Rainfall	Mean annual precipitation in inches
JanTemp	Mean January temperature in degrees Fahrenhiet
JulTemp	Mean July temperature in degrees Fahrenhiet
X.Over65	Percent of 1960 SMSA population which is 65 years of age or over
AvHHSize	Population per household
YrsEduc	Median school years completed for those over 25 in 1960 SMSA
X.GdHouse	Percent of housing units which are sound with all facilities
PopDens	Population per square mile in urbanized area in 1960
X.NWhite	Percent of 1960 urbanized area population which is non-white
WCollar	Percent employment in white-collar occupations in 1960 urbanized area
X.Poor	Percent families with income under \$3000 in 1960 urbanized area(cm)
HydCarb	Relative population potential of hydrocarbons
NOx	Relative population potential of oxides of nitrogen
SO2	Relative population potential of sulpher dioxides
RelHumid	Percent relative humidity, annual average at 1 PM
MortRate	Total age adjusted mortality rate, expressed as death per 100000 population.

Source

Alan Miller at http://users.bigpond.net.au/amiller/

References

Gunst, R.F. and Mason, R.L. (1980). Regression Analysis and its Application, Mercel Decker: New York, 370 – 371.

McDonald, G.C. and Schwing, R.C. (1973). Instabilities of regression coefficient estimates relating air pollution to mortality.Technometrics, 15, 463 – 482.

Examples

data(Pollute, package = "modelSampler")

tanujitdey/modelSampler documentation built on May 5, 2019, 11:01 p.m.